# Priors and Posteriors are almost the same

I have now estimated a model, where I get the following results:

parameters
prior mean post. mean 90% HPD interval prior pstdev

h 0.740 0.7412 0.7412 0.7413 norm 0.0100
sigma_c 1.353 1.3589 1.3588 1.3591 norm 0.0500
sigmal 2.400 2.3953 2.3952 2.3954 norm 0.0500
epsilonp 0.908 0.9043 0.9041 0.9045 beta 0.0500
epsilonw 0.737 0.7433 0.7430 0.7435 beta 0.0500
indp 0.469 0.4615 0.4612 0.4617 beta 0.0500
indw 0.763 0.7569 0.7566 0.7571 beta 0.0500
phi_p 1.500 1.4977 1.4975 1.4979 norm 0.0500
phi_w 1.500 1.5054 1.5052 1.5057 norm 0.0500
lambda 0.300 0.3014 0.3014 0.3015 beta 0.0100
phi 1.000 1.0068 1.0066 1.0070 norm 0.0500
phi_pi 1.500 1.4990 1.4975 1.5004 norm 0.5000
phi_y 0.500 0.5004 0.4999 0.5007 norm 0.1000
rhoii 0.750 0.7351 0.7346 0.7355 beta 0.1000
eta 1.200 1.2016 1.2014 1.2017 norm 0.1000
rhob 0.750 0.7532 0.7529 0.7534 beta 0.1000
rhoa 0.750 0.7493 0.7487 0.7497 beta 0.1000
rhog 0.750 0.7548 0.7543 0.7552 beta 0.1000
rhoinv 0.750 0.7391 0.7388 0.7394 beta 0.1000
rhop 0.750 0.7515 0.7513 0.7517 beta 0.1000
rhow 0.750 0.7575 0.7571 0.7577 beta 0.1000
rhoi 0.750 0.7428 0.7426 0.7431 beta 0.1000
rhoyf 0.750 0.7436 0.7429 0.7441 beta 0.1000
rhopif 0.750 0.7436 0.7432 0.7439 beta 0.1000

There is almost no difference between priors and posteriors and the posterior distribution is almost identical with the posterior mean. Is this a consequence from my assumption that the priors have a standard deviation of 0.05?

Did you check identification? You should be able to see this already in the mode_check plots as a horizontal likelihood.

The likelihood is horizontal for a few parameters, ie. three or four, but not for others.

Now I get the following:

ESTIMATION_CHECKS: There was an error in computing the likelihood for initial parameter values.
ESTIMATION_CHECKS: If this is not a problem with the setting of options (check the error message below),
ESTIMATION_CHECKS: you should try using the calibrated version of the model as starting values. To do
ESTIMATION_CHECKS: this, add an empty estimated_params_init-block with use_calibration option immediately before the estima
tion
ESTIMATION_CHECKS: command (and after the estimated_params-block so that it does not get overwritten):

The variance of the forecast error remains singular until the end of the sample

1. Try to eliminate the identification errors.
2. The error messages suggests stochastic singularity. What did you change?

Instead of a mat-file with all 7 variables separately, I declared a variable ts=dseries(DSGE_data.mat). This generated a dseries object, which was then used for estimation. But this time the estimation does not run and gives this error message.

That is strange. Please provide the codes to replicate the issue.

Please find attached the mod-file and the dataset:

DSGE_series_dat.mat (6.6 KB)
modelestim.mod (3.9 KB)